Preliminary experiments indicate that estrogen infusions effectively shorten prolonged bleeding time in uremic patients. The application's long-term objectives are to evaluate whether estrogen treatment is a rational alternative to available therapies for uremic bleeding tendency and to elucidate the pathogenesis of uremic bleeding by studying estrogen mechanism of action. Clinical information on the estrogen effect on uremic bleeding will be achieved by investigating whether single or repeated doses of estrogens are required to obtain a control of bleeding time and studying estrogen pharmacokinetics. Furthermore, the therapeutic effect of estrogen in 12 uremic patients with severe untractable hemorrhages will be evaluated in an open study. To elucidate estrogen mechanism of action, focus will be put on three parameters reported to be abnormal in uremic patients, which might contribute to the derangement of primary hemostasis and might respond favorably to estrogens. Red cells and their role in platelet-vessel wall interaction will be studied by measuring whole blood and plasma viscosity, red blood cell deformability, and platelet adhesion to rat subendothelium in well-defined flowing conditions. Metabolites of vascular prostacyclin will be measured in uremic plasma and in aortic segments from rats with experimental uremia. Antibody-mediated extraction/gas chromatography-negative ion chemical ionization mass spectrometry technique will be used for this purpose. The possible role of parathyroid hormone in uremic platelet dysfunction will be assessed by correlating plasma levels of parathyroid hormone or its fragments with bleeding times in 40 uremic patients and in 5 uremic patients before and after parathyroidectomy. The possibility that estrogens are effective in uremic patients because they modify levels or activity of the three above mentioned parameters will be investigated by comparing results obtained before and after appropriate estrogen treatment.